dipanshu231099 / inter-iit-2020-bosch-ds Goto Github PK

show the stats of the initial model in form of interactive graphs and also when ever a model is retrained show the new graphs.

We need to pass the list containing the augmentations in a particular order to a function in views.py from where they need to be passed to the functions defined in the train_model.py for the proper retraining and augmentations in the data.
Also there should be another parameter passed that decides weather to apply the augments to the entire data or test data or train+test data
P.S add a nice front end to the augmentations webpage

we need to have a nav bar in the base.html webpage in order to allow the user to either

1. Upload the Images along with its classes.
2. Upload the Images along with its classes and merge it with train and then apply the augmentations. also store it in a different .npy file.
3. Simply apply augmentations to the data and store it in a different .npy file
4. an option to retrain the data depending upon the above options.

I am working on the back end of this so if some one can work out the front end it will mostly consist of making 2-3 forms and passing there values it would be great

if the option to retrain is chosen then display the augmented and the non augmented images. One from each class.
or it would also be better if there is an option to check the augmented vs non augmented image before retraining the model.